NVIDIA Launches Open Agent Safety Platform to Secure Agents From Testing to Deployment
NVIDIA has introduced the Open Agent Safety Platform, featuring OpenShell open-source software and the Sentry reference system design. This platform aims to enhance AI security with full-stack governance and control for AI agents. OpenShell provides a secure runtime boundary, tracing actions and enforcing policies, while Sentry, an out-of-band watchdog on BlueField-4 DPUs, continuously monitors agent behavior and can quarantine agents in milliseconds if they exceed their boundaries. Anthropic is collaborating with NVIDIA on this platform.
This development means AI agents will operate with enhanced security measures, reducing the risk of unintended or harmful actions and increasing your confidence in delegating sensitive tasks to them.
Learn one new AI thing every day.
Daily Deck sends you seven plain-English cards like this every morning. Free.
Start free